Diagnostic accuracy of large language models in psychiatry.

INTRODUCTION: Medical decision-making is crucial for effective treatment, especially in psychiatry where diagnosis often relies on subjective patient reports and a lack of high-specificity symptoms. Artificial intelligence (AI), particularly Large Language Models (LLMs) like GPT, has emerged as a promising tool to enhance diagnostic accuracy in psychiatry. This comparative study explores the diagnostic capabilities of several AI models, including Aya, GPT-3.5, GPT-4, GPT-3.5 clinical assistant (CA), Nemotron, and Nemotron CA, using clinical cases from the DSM-5. METHODS: We curated 20 clinical cases from the DSM-5 Clinical Cases book, covering a wide range of psychiatric diagnoses. Four advanced AI models (GPT-3.5 Turbo, GPT-4, Aya, Nemotron) were tested using prompts to elicit detailed diagnoses and reasoning. The models' performances were evaluated based on accuracy and quality of reasoning, with additional analysis using the Retrieval Augmented Generation (RAG) methodology for models accessing the DSM-5 text. RESULTS: The AI models showed varied diagnostic accuracy, with GPT-3.5 and GPT-4 performing notably better than Aya and Nemotron in terms of both accuracy and reasoning quality. While models struggled with specific disorders such as cyclothymic and disruptive mood dysregulation disorders, others excelled, particularly in diagnosing psychotic and bipolar disorders. Statistical analysis highlighted significant differences in accuracy and reasoning, emphasizing the superiority of the GPT models. DISCUSSION: The application of AI in psychiatry offers potential improvements in diagnostic accuracy. The superior performance of the GPT models can be attributed to their advanced natural language processing capabilities and extensive training on diverse text data, enabling more effective interpretation of psychiatric language. However, models like Aya and Nemotron showed limitations in reasoning, indicating a need for further refinement in their training and application. CONCLUSION: AI holds significant promise for enhancing psychiatric diagnostics, with certain models demonstrating high potential in interpreting complex clinical descriptions accurately. Future research should focus on expanding the dataset and integrating multimodal data to further enhance the diagnostic capabilities of AI in psychiatry.

The association between increased fetal movements in the third trimester and perinatal outcomes; a systematic review and meta-analysis.

BACKGROUND: Fetal movement monitoring is one of the strategies used to assess the fetus's health. Until now, most studies focused on the decreased fetal movement and neonatal outcome, although this systematic review and meta-analysis is designed to assess the association between increased fetal movements (IFM) with perinatal outcomes. METHOD: The electronic databases including PubMed, Scopus, Web of Science, and EMBASE were systematically searched for studies investigating the perinatal outcome of women with increased fetal movements from inception to July 2023. Following that, a random-effect meta-analysis model was used to obtain the combined diagnostic and predictive parameters including perinatal mortality (still birth and early neonatal mortality), operative delivery, Apgar score, neonatal resuscitation at birth and NICU Admission. RESULTS: After the initial screening, seven studies examining the association between increased third trimester fetal movement and various perinatal outcomes were included. Meta-analysis revealed a significant reduction in the risk of cesarean delivery among patients with IFM compared to controls, suggesting a potential protective effect during childbirth. However, no statistically significant difference was observed in birth weight, small or large for gestational age births, neonatal intensive care unit admission, maternal age, umbilical cord around the neck, gestational diabetes mellitus, and hypertension, indicating that IFM may not be a major predictor of adverse perinatal outcomes or maternal conditions. Notably, IFM was significantly associated with a higher likelihood of labor induction. CONCLUSION: The findings suggest that IFM may have a protective effect against cesarean delivery. Additionally, IFM does not appear to be significantly associated with maternal age, umbilical cord around the neck, gestational diabetes mellitus and hypertension. However, the observed significant association with labor induction warrants further investigation.

The effect of COVID-19 and COVID-19 vaccination on serum anti-Mullerian hormone: A systematic review and meta-analysis.

OBJECTIVE: The current study aims to evaluate the impact of COVID-19 infection and vaccination on ovarian reserve by detecting the anti-Mullerian hormone (AMH) level. METHOD: PubMed, Embase, Web of Science, and Scopus has been searched for studies assessing the effect of COVID-19 infection and/or vaccination on AMH levels up to February 27, 2023. Based on PRISMA 2020 statement criteria, a systematic review and meta-analysis of included studies were performed. The studies' quality was assessed by the National Institute of Health (NIH) quality assessment tool. The standardized mean difference (MD) of the AMH level was used and the quantitative values of each study were pooled separately by using a random effect model. RESULTS: Out of 246 studies screened, 18 were included in the systematic review and 14 in the meta-analysis. Included studies were published between 2021 and 2022 and were conducted in different countries, including the USA (n = 3), China (n = 2), Russia (n = 2), Turkey (n = 5), Israel (n = 3), Czech (n = 2), and Spain (n = 1). Eight studies investigated the effect of SARS-CoV-2 infection on AMH levels, and ten studies investigated the possible effect of COVID-19 vaccination on AMH levels. The pooled analysis showed a statistically significant decrease in AMH levels after COVID-19 infection (SMD: -0.24; 95% CI: -0.36 to -0.11; I2 = 0%; p = .0003). Vaccination analysis showed a nonstatistically significant change in AMH levels after COVID-19 vaccination (SMD: -0.11; 95% CI: -0.25 to 0.04; I2 = 35%; p = .14). CONCLUSION: COVID-19 infection can result in ovarian reserve injury by reducing the AMH level but getting vaccinated against COVID-19 has no impact on the AMH level.

Prediction of intraperitoneal adhesions in repeated cesarean sections: A Systematic review and Meta-analysis.

OBJECTIVE: With the rate of repeated cesarean sections on the rise, intraperitoneal adhesions caused by repeated cesareans could give rise to maternal morbidity during delivery. As a result, it's critical to have the ability to predict adhesions. The current meta-analysis aims to determine if intraperitoneal adhesions are likely to be present based on the characteristics of the cesarean scar, striae gravidarum, and sliding sign. MATERIALS AND METHODS: We systematically searched electronicdatabases beforeretrieving articles up until October 13th, 2022 for analysis. After data extraction and literature screening, we first performed a quality assessment using the QUADAS-2 score system. Following that, a bivariate random-effect meta-analysis model was used to obtain the combined diagnostic and predictive values. To pinpoint the origins of heterogeneity, we conducted a subgroup analysis. Fagan's Nomogram was used to validate the clinical utility. Sensitivity analysis was used to gauge the reliability of each included study, and Egger's test and funnel plot asymmetry was used to investigate publication bias. RESULTS: 25 studies totaling 1840 patients with intra-abdominal adhesions and 2501 controls without adhesions were included in the systematic review. Diagnostic values from 8 studies regarding skin characteristics were combined, and the results for depressed scar showed: sensitivity[95 %CI] = 0.38[0.34-0.42]; Specificity[95 %CI] = 0.88[0.85-0.90]; DOR[95 %CI] = 4.78[2.50-9.13]; AUC = 0.65. Negative sliding sign from 7 studies, although not showing a diagnostic difference between cases and controls, had excellent predictive values: sensitivity[95 %CI] = 0.71[0.65-0.77]; Specificity[95 %CI] = 0.87[0.85-0.89]; DOR[95 %CI] = 6.88[0.6-78.9]; AUC = 0.77. Subgroup analysis illustrated non-Turkish studies to reveal more significant associations than Turkish studies. CONCLUSION: Our meta-analysis found that the occurrence of adhesions can be predicted by the characteristics of abdominal wounds, particularly depressed scar, and scar width,as well as a negative sliding sign following a previous cesarean section.